Painters don't tell. Painters show. David Hockney's breakthrough work on the use made by the old masters of mirrors and lenses is becoming well-known. Though there are carpers and hecklers, it has leapt from hunch, through theory, to accepted fact, in a remarkably short time. His basic idea, that lenses - effectively, photography without the chemical fixing - spread throughout European painting from the Renaissance onwards, alters for ever our own perception of some of the greatest artists in history. But if you recoil from it, Hockney doesn't harangue or argue. He shows.
His first way of amassing the evidence was to do something you or I wouldn't dream of, which is to try using lens techniques for himself. Two years ago, Hockney was at the National Gallery's Ingrès show in London. He was gripped with curiosity at the speed, accuracy and odd certainties of Ingrès's extraordinary fast portrait sketches done in Rome (many of them of British tourists). How had he done it? Hockney was suspicious. There were tiny hints in the pencil marks, things non-artists would never have spotted, that reminded him of Andy Warhol's drawings, actually tracings. He thought Ingrès might have used a newly invented optical device called a camera lucida. Hockney found one, taught himself to use it, and was soon also doing very fast, very accurate pencil portraits with its help.
Then the fun began. The more paintings Hockney studied, from the 1400s to the impressionists, the more evidence he thought he saw of other lenses being used. This is not to say that the greats of Western art - Durer, Holbein, van Eyck, Raphael, Caravaggio, Velazquez, Hals and many more - were 'cheating'. Why is using a lens cheating, any more than using a plumb-line or a rule-book of mathematical perspective solutions? But they were certainly being helped with... well, with photography.
Hockney has always enjoyed a row, and is savouring this one. When I met him in his London studio, he was flinging out ideas like showers of sparks from a Catherine wheel. 'The photograph is far, far older than we think. That kind of image is older. It's just that they didn't have the chemical fixative until the nineteenth century,' he says. 'It frees us. It makes the artists of the past much, much closer. They are not demi-gods way up there. They are marvellous artists but their techniques have a great deal to teach the artists of today.'
The idea that Vermeer used lenses - particularly a camera obscura for his famous view of Delft, but also for his interiors - has become reasonably familiar. And the circumstantial evidence for Caravaggio, who did no preliminary drawings, worked in a darkened cellar, and used the back of his brush to make very quick marks on a ground of wet paint, is extremely strong. Indeed, once you look at his lighting and foreshortening, which is very 'photographic', it's obvious. But Hockney has pushed the argument far, far, further, spreading the use of lenses across four centuries and scores of other reputations. It is the biggest shift in our understanding of the old masters I can think of.
Once Hockney was on this track, he created a massive gallery of photos and postcards of paintings on a wall of his Californian studio, running from the 1300s to the last century, divided into northern and southern European, along which breakthroughs in realism become strikingly obvious. Each, he thinks, corresponds to a new technological breakthrough in lens-making, with the first around 1420. That is well-known and much discussed. But why did it happen? Hockney giggles in derision: 'Oh, they say, suddenly in 1420 everyone could draw better. From that moment you never see a badly drawn basket again in Western art. They are suddenly all perfectly woven, in perfect perspective.' No, he says, the answer is the new lenses that spread from the Low Countries to Florence and then the rest of Italy. And so it continued.
The use of lenses to produce living images to gape at is well-known and goes back to the Middle Ages. There are plenty of written accounts. It is very easy to do. While Hockney was being photographed, his assistant showed me how. But what we hadn't known is how much lenses, particularly concave mirrors, were used to help draw and paint so realistically.
The same story applies to moving ones. Lenses, of course, show movement: the projections enjoyed by artists hundreds of years ago, Hockney told me, 'are far better than high-definition television. In clarity and colour, they are amazing, they are ravishingly beautiful.' But in the days of van Eyck or Caravaggio, they could not be preserved, except by drawing, a kind of note-taking. 'They were living projections, meaning, if they are flowers, the flowers will die; if it's a person, the person will move. But there's no doubt that they saw colour images, optical projections which look like paintings. And they made paintings which look like optical projections.' He chortles with delight at the thought. The old masters saw moving colour pictures - proto-television. It brings them closer, doesn't it?
After he had taught himself with Ingrès's lens, and created his wall of images, and was touring galleries around the world, note-taking, Hockney pursued the theory with art historians and friends, by fax - above all with Martin Kemp, the professor of art history at Oxford University, and Charles Falco, an optical scientist. From everywhere, the evidence accumulated. A mere artist, a hand and brush and pencil man, seemed to have spotted what the combined intellects of academic art history had missed.
But, as I said, what an artist does is to show. You can see Hockney's big idea for yourself now, in a remarkable book, which includes his own camera lucida drawings; page after page of gorgeously rich visual evidence; and even those collected faxes. Whereas most good art books deepen your understanding of one painter or a group of artists, this one is far more ambitious. I have been in quite a few galleries since meeting Hockney and poring over the book, and it changes the way you see hundreds of paintings he never mentions. I spent a happy couple of hours recently in the Musée des Beaux-Arts in Brussels seeing work after work afresh. Not spoiled, just differently.
What is not obvious from the book, but was once Hockney started talking, was how this big idea develops into the modern era and how it connects smoothly to Hockney's other, more conventional works, all those paintings, photographs and writings. It sounds, at first hearing, a complex theory. But here is the simple version. Think of three periods of art history, all defined by photography.
First, the use of lenses from the Renaissance to the mid-nineteenth century. A trick of the artist's trade, not much talked about to laymen, but exposed through the Hockney sleuthing. (There is little written evidence, he argues, because lenses were seen as dangerous things, close to witchcraft and partly controlled by the church, in the early days; later on, which artist wants to show how he does what he does?)
Second, the arrival of chemically fixed photography, which immediately threatens the painters' trade union. You can get a snap, why bother with an oil? Artists respond by either trying to turn back to the pre-lens era (the pre-Raphaelites in England) or by emphasising a deliberately rugged, anti-photographic style of painting (Cézanne, Van Gogh, and eventually Picasso).
But then what happens? Answer: the computer, and the third phase, through which we are living today. In 1975, the great French photographer, Cartier-Bresson told Hockney he was giving up photography. Now, Hockney says, he realises the significance of the date. 'Cartier-Bresson starts in 1925, the time of the arrival of the 35mm camera and faster film, so you can carry a camera around with you - you don't need a tripod. In about 1975 the computer arrives to change photography and, in a sense, the purity of the Cartier-Bresson period.'
The computer, through technology such as Adobe's Photoshop, allows the manipulation of the photographic image. Chemical photography gives way to digital technology, which can be infinitely manipulated. And what does manipulation mean, asks Hockney? It means changing by hand, making by hand - it means drawing. 'Photographs are going to be more and more made by drawing, so the photograph has really lost its veracity.'
And here, triumphantly, the Hockney case that lenses were used by the great masters comes full circle, returning us in 2001 to the revival of drawing, another life-long passion. 'For 400, nearly 500 years, the hand was involved with the camera. Artists were in it, using the lens. Then for 160 years, you had chemical photography. But that has now come to an end, and with digital photography, you have got the hand back in the camera. This is why' - and here he flicks open a glossy magazine to demonstrate the point - '... photography is changing and actually moving back towards drawing and painting.'
And Hockney lights another cigarette. And leans back. And laughs. As well he might.
Secret Knowledge: Rediscovering the Lost Techniques of the Old Masters by David Hockney, is published by Thames & Hudson, £35. An Omnibus special is screened on BBC2 next Saturday.